ToKen: Task Decomposition and Knowledge Infusion for Few-Shot Hate Speech Detection
This work addresses the challenge of detecting hate speech with limited annotated data, which is a problem for social media platforms and content moderators, though it appears incremental as it builds on existing few-shot and knowledge infusion methods.
The paper tackled hate speech detection as a few-shot learning problem by decomposing the task into constituent parts and infusing knowledge from reasoning datasets, achieving a 17.83% absolute gain over the baseline in the 16-shot case.
Hate speech detection is complex; it relies on commonsense reasoning, knowledge of stereotypes, and an understanding of social nuance that differs from one culture to the next. It is also difficult to collect a large-scale hate speech annotated dataset. In this work, we frame this problem as a few-shot learning task, and show significant gains with decomposing the task into its "constituent" parts. In addition, we see that infusing knowledge from reasoning datasets (e.g. Atomic2020) improves the performance even further. Moreover, we observe that the trained models generalize to out-of-distribution datasets, showing the superiority of task decomposition and knowledge infusion compared to previously used methods. Concretely, our method outperforms the baseline by 17.83% absolute gain in the 16-shot case.